diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp.hpp
index 295b2c21b5..e42a3f2045 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp.hpp
@@ -297,6 +297,9 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
             {
                 add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instances(
                     op_ptrs);
+
+                add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instances(
+                    op_ptrs);
             }
 #endif
 #ifdef CK_ENABLE_FP16
@@ -306,6 +309,9 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
             {
                 add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instances(
                     op_ptrs);
+
+                add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instances(
+                    op_ptrs);
             }
 #endif
         }
@@ -322,6 +328,9 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
             {
                 add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
                     op_ptrs);
+
+                add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+                    op_ptrs);
             }
 #endif
 #ifdef CK_ENABLE_FP16
@@ -331,6 +340,9 @@ struct DeviceOperationInstanceFactory<ck::tensor_operation::device::DeviceGroupe
             {
                 add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_ndhwgk_f16_instances(
                     op_ptrs);
+
+                add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instances(
+                    op_ptrs);
             }
 #endif
         }
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp_wmma_cshufflev3.inc b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp_wmma_cshufflev3.inc
index e2ad6df07e..90e84635bd 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp_wmma_cshufflev3.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_bias_bnorm_clamp_wmma_cshufflev3.inc
@@ -24,6 +24,21 @@ void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhw
                                                         PassThrough,
                                                         BiasNormalizeInInferClamp>>>& instances);
 
+void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instances(
+    std::vector<
+        std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                        NHWGC,
+                                                        GKYXC,
+                                                        Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                                        NHWGK,
+                                                        BF16,
+                                                        BF16,
+                                                        Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                                        BF16,
+                                                        PassThrough,
+                                                        PassThrough,
+                                                        BiasNormalizeInInferClamp>>>& instances);
+
 void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
     std::vector<std::unique_ptr<
         DeviceGroupedConvFwdMultipleABD<3,
@@ -38,6 +53,21 @@ void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_n
                                         PassThrough,
                                         PassThrough,
                                         BiasNormalizeInInferClamp>>>& instances);
+
+void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+    std::vector<std::unique_ptr<
+        DeviceGroupedConvFwdMultipleABD<3,
+                                        NDHWGC,
+                                        GKZYXC,
+                                        Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                        NDHWGK,
+                                        BF16,
+                                        BF16,
+                                        Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                        BF16,
+                                        PassThrough,
+                                        PassThrough,
+                                        BiasNormalizeInInferClamp>>>& instances);
 #endif
 
 #ifdef CK_ENABLE_FP16
@@ -56,6 +86,21 @@ void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhw
                                                         PassThrough,
                                                         BiasNormalizeInInferClamp>>>& instances);
 
+void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instances(
+    std::vector<
+        std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                        NHWGC,
+                                                        GKYXC,
+                                                        Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                                        NHWGK,
+                                                        F16,
+                                                        F16,
+                                                        Tuple<F16, F16, F16, F16, F16>,
+                                                        F16,
+                                                        PassThrough,
+                                                        PassThrough,
+                                                        BiasNormalizeInInferClamp>>>& instances);
+
 void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_ndhwgk_f16_instances(
     std::vector<std::unique_ptr<
         DeviceGroupedConvFwdMultipleABD<3,
@@ -70,6 +115,21 @@ void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_ndhwgc_gkzyxc_n
                                         PassThrough,
                                         PassThrough,
                                         BiasNormalizeInInferClamp>>>& instances);
+
+void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instances(
+    std::vector<std::unique_ptr<
+        DeviceGroupedConvFwdMultipleABD<3,
+                                        NDHWGC,
+                                        GKZYXC,
+                                        Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                        NDHWGK,
+                                        F16,
+                                        F16,
+                                        Tuple<F16, F16, F16, F16, F16>,
+                                        F16,
+                                        PassThrough,
+                                        PassThrough,
+                                        BiasNormalizeInInferClamp>>>& instances);
 #endif
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/CMakeLists.txt
index d089663f37..1f381f5f7d 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/CMakeLists.txt
@@ -328,6 +328,8 @@ generate_sharded_instantiations(
 add_instance_library(device_grouped_conv2d_fwd_bias_bnorm_clamp_instance
    wmma/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
    wmma/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+   wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+   wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instance.cpp
    ${GROUPED_CONV2D_FWD_BIAS_BNORM_CLAMP}
 )
 
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
new file mode 100644
index 0000000000..f2729fe0e4
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
@@ -0,0 +1,44 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_cshufflev3_large_tensor_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using BiasNormalizeInInferClamp = ck::tensor_operation::element_wise::BiasNormalizeInInferClamp;
+
+void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_bf16_instances(
+    std::vector<
+        std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                        NHWGC,
+                                                        GKYXC,
+                                                        Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                                        NHWGK,
+                                                        BF16,
+                                                        BF16,
+                                                        Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                                        BF16,
+                                                        PassThrough,
+                                                        PassThrough,
+                                                        BiasNormalizeInInferClamp>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_wmma_large_tensor_bf16_generic_instances<
+                                       2,
+                                       NHWGC,
+                                       GKYXC,
+                                       Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                       NHWGK,
+                                       ConvFwdDefault,
+                                       Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                       BiasNormalizeInInferClamp>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instance.cpp
new file mode 100644
index 0000000000..7be4be2f1e
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instance.cpp
@@ -0,0 +1,44 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_cshufflev3_large_tensor_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using BiasNormalizeInInferClamp = ck::tensor_operation::element_wise::BiasNormalizeInInferClamp;
+
+void add_device_grouped_conv2d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_nhwgc_gkyxc_nhwgk_f16_instances(
+    std::vector<
+        std::unique_ptr<DeviceGroupedConvFwdMultipleABD<2,
+                                                        NHWGC,
+                                                        GKYXC,
+                                                        Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                                        NHWGK,
+                                                        F16,
+                                                        F16,
+                                                        Tuple<F16, F16, F16, F16, F16>,
+                                                        F16,
+                                                        PassThrough,
+                                                        PassThrough,
+                                                        BiasNormalizeInInferClamp>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_wmma_large_tensor_f16_generic_instances<
+                                       2,
+                                       NHWGC,
+                                       GKYXC,
+                                       Tuple<NHWGK, NHWGK, NHWGK, NHWGK, NHWGK>,
+                                       NHWGK,
+                                       ConvFwdDefault,
+                                       Tuple<F16, F16, F16, F16, F16>,
+                                       BiasNormalizeInInferClamp>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/CMakeLists.txt b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/CMakeLists.txt
index dc759cbb54..f54588991f 100644
--- a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/CMakeLists.txt
@@ -309,6 +309,8 @@ generate_sharded_instantiations(
 add_instance_library(device_grouped_conv3d_fwd_bias_bnorm_clamp_instance
    wmma/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
    wmma/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+   wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
+   wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp
    ${GROUPED_CONV3D_FWD_BIAS_BNORM_CLAMP}
 )
 
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
new file mode 100644
index 0000000000..4a9c68b2d3
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instance.cpp
@@ -0,0 +1,44 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_cshufflev3_large_tensor_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using BiasNormalizeInInferClamp = ck::tensor_operation::element_wise::BiasNormalizeInInferClamp;
+
+void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_bf16_instances(
+    std::vector<std::unique_ptr<
+        DeviceGroupedConvFwdMultipleABD<3,
+                                        NDHWGC,
+                                        GKZYXC,
+                                        Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                        NDHWGK,
+                                        BF16,
+                                        BF16,
+                                        Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                        BF16,
+                                        PassThrough,
+                                        PassThrough,
+                                        BiasNormalizeInInferClamp>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_wmma_large_tensor_bf16_generic_instances<
+                                       3,
+                                       NDHWGC,
+                                       GKZYXC,
+                                       Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                       NDHWGK,
+                                       ConvFwdDefault,
+                                       Tuple<BF16, BF16, BF16, BF16, BF16>,
+                                       BiasNormalizeInInferClamp>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck
diff --git a/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp
new file mode 100644
index 0000000000..92c86b8df0
--- /dev/null
+++ b/library/src/tensor_operation_instance/gpu/grouped_conv3d_fwd_bias_bnorm_clamp/wmma/large_tensor/device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instance.cpp
@@ -0,0 +1,44 @@
+// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
+// SPDX-License-Identifier: MIT
+
+#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
+#include "ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_wmma_cshufflev3_large_tensor_instance.hpp"
+
+namespace ck {
+namespace tensor_operation {
+namespace device {
+namespace instance {
+
+using BiasNormalizeInInferClamp = ck::tensor_operation::element_wise::BiasNormalizeInInferClamp;
+
+void add_device_grouped_conv3d_fwd_bias_bn_clamp_wmma_cshufflev3_large_tensor_ndhwgc_gkzyxc_ndhwgk_f16_instances(
+    std::vector<std::unique_ptr<
+        DeviceGroupedConvFwdMultipleABD<3,
+                                        NDHWGC,
+                                        GKZYXC,
+                                        Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                        NDHWGK,
+                                        F16,
+                                        F16,
+                                        Tuple<F16, F16, F16, F16, F16>,
+                                        F16,
+                                        PassThrough,
+                                        PassThrough,
+                                        BiasNormalizeInInferClamp>>>& instances)
+{
+    add_device_operation_instances(instances,
+                                   device_grouped_conv_fwd_wmma_large_tensor_f16_generic_instances<
+                                       3,
+                                       NDHWGC,
+                                       GKZYXC,
+                                       Tuple<NDHWGK, NDHWGK, NDHWGK, NDHWGK, NDHWGK>,
+                                       NDHWGK,
+                                       ConvFwdDefault,
+                                       Tuple<F16, F16, F16, F16, F16>,
+                                       BiasNormalizeInInferClamp>{});
+}
+
+} // namespace instance
+} // namespace device
+} // namespace tensor_operation
+} // namespace ck